How we test

This page states exactly what DeAIze measures, how a rewrite is accepted, and how to read the numbers we publish. It is written so that anyone can check our claims against their own text.

1. What a detector score is

Detectors do not detect authorship; they detect statistical regularity. Machine-generated text tends to be more predictable than human writing — lower perplexity, flatter sentence rhythm, fewer lexical surprises. Every score you see, ours included, is a probability estimate from that signal, not a measurement of who wrote the text.

2. Sentence-level scoring

A single document score tells you nothing about where the problem is. Rewrite only the sentences that read machine-made and leave the ones that are already yours. Our own check shows you which is which before you spend a word. The per-sentence scores are what the rewrite loop acts on, and they are what the workbench highlights.

3. The closed loop

In seconds you get writing that reads like a person wrote it, with your meaning intact and the change gated by semantic similarity — not a synonym swap. A rewrite candidate is accepted only when semantic similarity to the original clears a threshold, which is why meaning is preserved and why the tool reports an "auto-refund on no-op" when it cannot improve a passage rather than inventing a change.

4. Detector cross-checks

A prediction from our own model is not evidence. Where a detector API is configured, the verify step sends the rewritten text to the real detector and records that result. The count of those runs is published on the statistics page so the number of third-party checks behind our figures is visible. A first-party figure is not a cross-check: only runs actually sent to a detector API are counted there.

5. What our own pattern score means

It is our own model's estimate of how heavily a passage leans on patterns that read machine-made — a diagnostic for tracking one revision, not an official score from any third-party detector. When we quote a third-party number (GPTZero, Turnitin, Copyleaks, Originality.ai, Sapling) we say so and, where we can, record the run as a cross-check.

First-party figures. Any figure DeAIze reports is our own, produced by our own model, and where a detector API could be called it is recorded as a cross-check under §4. The detector vendors do not publish or audit our numbers, and no figure here is a guarantee for any individual document. Turnitin's AI writing indicator has no public scoring API, so there is no live Turnitin readout anywhere on this site. We date our figures and we do not present them as an independent benchmark.

6. Limits and disclosure

• Detector scores vary by version, language, text length and domain. A number without a date and a detector version is not reproducible, so we date our published figures.
• Non-English performance is weaker than English, and short texts are noisier than long ones.
• When we publish a benchmark it carries its sample size, test date and detector versions. First-party figures are labelled as first-party rather than as an audited benchmark, because that is what they are.
• An automated score is an estimate, not proof. Editing your own draft is always legitimate; misrepresenting someone else's work never is.

Corrections

If any figure here cannot be reproduced, tell us and we will correct it: [email protected].